Routing Control Method, Apparatus and System

1. A routing control method, comprising: providing an apparatus comprising a minimum cost path search processor and a predicted utilization transition data storage storing predicted utilization transition data in future for each resource in a network; said minimum cost path search processor reading out predicted utilization transition data associated with a received connection set-up request from said predicted utilization transition data storage; said minimum cost path search processor selecting a resource satisfying a predetermined condition; and said minimum cost path search processor detecting an arrival of a communication request for a large amount of traffic whose bandwidth or holding time is greater than a predetermined value, based on said predicted utilization transition data, and wherein said selecting comprises: obtaining a cost of each said resource for a communication request for traffic other than said large amount of traffic based on said predicted utilization transition data relating to said arrival of said communication request for said large amount of traffic to select a first resource satisfying a second predetermined condition including a condition that a cost of said first resource is other than a minimum cost; and obtaining a cost of each said resource for said communication request for said large amount of traffic by using current utilization status data for each resource of said network, which is stored in a network context data storage, or by using static attribute information stored in a network data storage to select a second resource satisfying a third predetermined condition including a condition that a cost of said second resource is said minimum cost.

2. The routing control method as set forth in claim 1 , wherein said selecting comprises identifying a value representing the strictest predicted utilization status in a period associated with said connection set-up request, in said read predicted utilization transition data.

3. The routing control method as set forth in claim 2 , wherein said period associated with said connection set-up request is a predicted communication holding period.

4. The routing control method as set forth in claim 1 , said apparatus further comprising a traffic transition predictor, and said routing control method further comprising: said traffic transition predictor generating predicted utilization transition data in future for each said resource of said network from past utilization status data for said each resource of said network, which is stored in said network context data storage, and storing said predicted utilization transition data into said predicted utilization transition data storage.

5. The routing control method as set forth in claim 1 , wherein said predicted utilization transition data storage stores said predicted utilization transition data and identification information of a server associated with said predicted utilization transition data or identification information of an edge router associated with said predicted utilization transition data, and said reading comprises identifying said predicted utilization transition data applied at least according to a connection destination address of said connection set-up request.

6. The routing control method as set forth in claim 5 , wherein said identifying comprises: detecting that identification information of a server identified from a connection destination address of said connection set-up request, or identification information of an edge router identified from a connection destination address of said connection set-up request is registered for a specific resource in association with said predicted utilization transition data in said predicted utilization transition data storage; and identifying another predetermined data for said specific resource as said predicted utilization transition data without using the corresponding predicted utilization transition data.

7. The routing control method as set forth in claim 5 , wherein said identifying comprises: detecting that identification information of a server identified from a connection destination address of said connection set-up request, or identification information of an edge router identified from a connection destination address of said connection set-up request is not registered for a specific resource in association with said predicted utilization transition data in said predicted utilization transition data storage; and identifying said predicted utilization transition data for said specific resource.

8. The routing control method as set forth in claim 6 , wherein said another predetermined data is one of current utilization status data, data representing no load, and static attribute information.

9. The routing control method as set forth in claim 1 , said apparatus further comprising a traffic transition predictor and a change pattern data storage, and said routing control method further comprising: said traffic transition predictor generating first predicted utilization transition data for each said resource of said network by using at least one of periodic change pattern data of a utilization status, non-periodic event change pattern data of said utilization status, and long-term change pattern data of said utilization status, which are generated in advance for each said resource of said network and stored in said change pattern data storage.

10. The routing control method as set forth in claim 9 , further comprising: said traffic transition predictor generating second predicted utilization transition data by modifying the generated first predicted utilization transition data so as to be identical with current utilization status data for each said resource of said network, which is stored in said network context data storage.

11. The routing control method as set forth in claim 10 , further comprising: said traffic transition predictor generating third predicted utilization transition data by modifying the generated second predicted utilization transition data in accordance with a physical upper limit value of each said resource of said network.

12. The routing control method as set forth in claim 1 , further comprising: forcibly rejecting a new path setting request using a resource, while said resource is judged as being unstable based on a difference between a current value identified from said predicted utilization transition data and an actual value of current utilization status data.

13. The routing control method as set forth in claim 1 , further comprising: detecting that a resource is judged as being unstable based on a difference between a current value identified from said predicted utilization transition data and an actual value of current utilization status data; and carrying out at least one of forcibly disconnecting a path already using said resource being unstable, switching traffic to another stable path, and discarding packets flowing through a path already using said resource being unstable, and wherein said carrying out is continuously executed until said resource is judged as being stable.

14. A routing control apparatus, comprising: a predicted utilization transition data storage device storing predicted utilization transition data in future for each resource in a network; a first unit to read out predicted utilization transition data associated with a received connection set-up request from said predicted utilization transition data storage device; a selection unit to select a route comprising resources satisfying a predetermined condition based on the read predicted utilization transition data; and a second unit to detect an arrival of a communication request for a large amount of traffic whose bandwidth or holding time is greater than a predetermined value, based on said predicted utilization transition data, and wherein said selection unit obtains a cost of each said resource for a communication request for traffic other than said large amount of traffic based on said predicted utilization transition data relating to said arrival of said communication request for said large amount of traffic to select a first resource satisfying a second predetermined condition including a condition that a cost of said first resource is other than a minimum cost; and said selection unit obtains a cost of each said resource for said communication request for said large amount of traffic by using current utilization status data for each resource of said network, which is stored in a network context data storage, or by using static attribute information stored in a network data storage to select a second resource satisfying a third predetermined condition including a condition that a cost of said second resource is said minimum cost.

15. A layered type routing control system, comprising: said routing control apparatus as set forth in claim 14 for each subnetwork; a second routing control apparatus that carries out a processing to control a route between subnetworks by managing said routing control apparatuses.

16. The layered type routing control system as set forth in claim 15 , wherein a computational resource is allocated to said routing control apparatuses and said second routing control apparatus in accordance with a rate determined from a number N of all nodes to be controlled and managed and a number M of subnetworks.

17. A routing control method, comprising: providing an apparatus comprising a request receiver, a path search processor, a predicted utilization transition data storage storing predicted utilization transition data in future for each resource in a network, a network data storage and a network context data storage; said request receiver receiving a connection set-up request; and said path search processor selecting a resource selection processing for said connection set-up request among a first processing, a second processing and a third processing, and wherein said first processing executed by said path search processor comprises: reading out predicted utilization transition data associated with the received connection set-up request from said predicted utilization transition data storage, and selecting a resource satisfying a first predetermined condition based on the read predicted utilization transition data, said second processing executed by said path search processor comprises: selecting a first resource satisfying a second predetermined condition by using data concerning a static attribute for each resource of said network and including a condition that a cost of said first resource is other than a minimum cost, said data concerning said static attribute stored in said network data storage, and said third processing executed by said path search processor comprises: selecting a second resource satisfying a third predetermined condition by using current utilization status data for each resource of said network and including a condition that a cost of said second resource is said minimum cost, said current utilization status data stored in said network context data storage.

18. The routing control method as set forth in 17 , wherein said selecting said resource selection processing is executed when generating a conditional network topology graph corresponding to said received connection set-up request, or when searching a network topology graph for a path corresponding to said received set-up connection set-up request.

19. The routing control method as set forth in 17 , wherein said selecting said resource selection processing is carried out when generating a conditional network topology graph, and said first, second or third predetermined condition comprises a condition in which a maximum value of a utilization rate in a period associated with said connection set-up request is equal to or less than a threshold, when said predetermined utilization transition data is said utilization rate, or a condition in which a minimum value of an available bandwidth in said period associated with said connection set-up request is equal to or greater than a threshold, when said predetermined utilization transition data is said available bandwidth.

20. The routing control method as set forth in 17 , wherein said selecting said resource selection processing is carried out when generating a conditional network topology graph, and said predetermined condition comprises a condition in which a difference between a current value identified from said predicted utilization transition data and an actual value of current utilization status data or a link cost that is a time integral value of said difference is within a threshold range.

21. The routing control method as set forth in claim 17 , wherein said selecting said resource selection processing is carried out when carrying out a path search, and said predetermined condition comprises a condition in which a minimum cost path is searched by using, as a cost, a delay time converted from a utilization rate in said predicted utilization transition data when carrying out a minimum delay path search taking into account a predicted transition.

22. The routing control method as set forth in claim 17 , wherein said selecting said resource selection processing is carried out when carrying out a path search, and said predetermined condition comprises a condition in which a minimum cost path is searched by using, as a link cost, a difference between a current traffic volume derived from said predicted utilization transition data and an actual current traffic volume, or a time integral value of said difference, when carrying out a maximum stability path search taking into account a predicted transition.